Unclear What the compact edit result changes in what you see when a subagent edits a file is not stated.
A subagent is a helper that Claude Code starts to work on part of a task on its own. When a subagent uses the Edit tool, the tool's result can now include a compact summary of the edit, plus a check that the summary still matches the file. The check reads the file again and compares a SHA-256 fingerprint (a short code computed from the file's exact contents) with the one taken at edit time.
The summary is attached only when all of these hold:
- The edit was made by a subagent, not the main conversation.
- A session check passes.
- A remotely controlled setting is switched on. It is off unless switched on remotely.
The setting is skipped if the request specifically asks for a small result. In that case the summary is built anyway. Neither the summary nor the check is built when the file or the edited text is large.
This lets a subagent's edits be reported back to the main conversation more cheaply. Until the setting is switched on for an account, edits behave as before.
tengu_patient_meteor Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.295: off
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.295. It isn't a statement about your account. What a flag value here can and cannot tell you
What the compact edit result changes in what you see when a subagent edits a file is not stated.
New in this build: tengu_patient_meteor